Protocol for AI agents to work together
The simplest protocol for AI agents to work together
It's not enough for AI agents to get smarter. They need to be able to work together, to interact, negotiate, and reach decisions — just like people.
IBM designed its Agent Communication Protocol (ACP) to make it easier for AI agents to collaborate and solve complex real-world tasks. ACP powers the open-source platform BeeAI which IBM recently contributed to the nonprofit Linux Foundation. ACP allows developers to orchestrate LLM agents with greater flexibility and zero vendor lock-in.
ACP versus MCP
ACP and Anthropic's Model Context Protocol (MCP) were built to support different layers of the stack, and can be used together in any multi-agent orchestration system, including BeeAI. ACP connects agents to agents; MCP connects agents to their tools and knowledge.
In a multi-agent system, a pair of agents could use MCP to call tools for gathering market data and running simulations, and ACP to compare their results and make a recommendation.
ACP uses a RESTful architecture implemented over HTTP, supporting both synchronous and asynchronous agent interactions, while MCP uses JSON-RPC.
The BeeAI team showed off an early version of ACP at the AI Dev 25 conference in March. A few weeks later, Google unveiled its own agent-to-agent protocol, A2A.
References:
https://www.linkedin.com/pulse/theres-new-agent-protocol-town-ibm-research-enjne/
https://youtu.be/Nzaq2S1EpLY
https://research.ibm.com/blog/agent-communication-protocol-ai
Comments
Post a Comment